Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
Beware of old silver-mica capacitors




 
The AM Forum
October 22, 2019, 05:20:02 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Beware of old silver-mica capacitors  (Read 24014 times)
0 Members and 1 Guest are viewing this topic.
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4590



« on: June 15, 2005, 10:30:40 AM »

Do you have a receiver with poor sensitivity that just won't align properly, or perhaps a transmitter with low output or an unstable oscillator?  You might be seeing the effects of bad silver-mica capacitors.

Apparently silver can migrate thru the mica insulation forming little fingers (dendrites) which can intermittently short the capacitor, cause capacitance changes, and/or affect the capacitor's Q.

Lately I've run into a rash of bad silver-micas.  You can recognize them from the square red, green, or brown molded cases with a 6-dot color code.  For example, while restoring a SX-42, the IF sensitivity was very low and none of the IF transformers would exhibit a sharp peak during alignment.  I replaced all of the postage-stamp style micas located inside the IF xformers and got nice sharp peaks as well as hugely increased sensitivity.

I looked at some of the old capacitors on my HP digital R-L-C bridge and the Q factor (which reads .001 or 0 on new, good caps) was reading 1.2 or so - and a 220 pF cap read 460 pF!  No wonder the thing wouldn't align properly.

This isn't a fun or easy job, but often the IF can cover can be removed from the top without having to remove the wires from the bottom of the chassis.  Be EXTREMELY careful of the thin Litz wire coming from the coil windings.  If you break one, wrap several turns of the wire around a clipped-off resistor lead and solder it well, then solder that to the terminal where it broke off.  If the Litz wire breaks off near the coil, you can almost always pull off a turn or so of wire without negative effect on the coil.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
W1GFH
Guest
« Reply #1 on: June 15, 2005, 11:53:50 AM »

Here, here! The VFO in my Ranger was hopping all over the place until I changed out the silver micas in the VFO box.


On another subject....when can we look forward to reading JOHNNY NOVICE'S GUIDE TO THE 10 BEST (AND WORST) AM TRANSMITTERS....?
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4590



« Reply #2 on: June 15, 2005, 12:08:56 PM »

Unfortunately, I don't have that much experience with xmitters.  The only AM xmitters I've had are a Viking II, a Valiant, a Cheyenne, and I just picked up a Ranger.  Not big enough statistical sample to draw any conclusions from, I'm afraid!

73 John
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Jim, W5JO
Member

Offline Offline

Posts: 2449


« Reply #3 on: June 15, 2005, 12:29:13 PM »

This is a subject that deserves some attention JN.  I have discovered several bad silver micas in receivers, most notably my HRO 60.  It has 510 pf silver micas across the input and output of the IF cans and when it was first turned on the sensitivity was fine, but as the circuits warmed the sensistivity would go away.

Finally I did stage gain measurments and found the second IF sensitivity was drifting.  It took me a while to notice that the problem was within the IF can, and as you described the peak would flatten out to almost nonexistant.

According to the old color code books the silver mica had a black dot for the first dimple on the JAN and AWS codes  Other codes didn't identify the diaelectric.  As far as I know there are two kinds of these things.  One is the silver mica and the other is paper.  We all know what moisture can do to paper.

It takes some common sense and troubleshooting skills to identify these problems, but I suggest a little patience and sometimes just rebuilding the circuit if you don't find it by normal procedures.  Good find.
Logged
W1GFH
Guest
« Reply #4 on: June 15, 2005, 07:37:29 PM »

Quote from: w3jn
Unfortunately, I don't have that much experience with xmitters.  The only AM xmitters I've had are a Viking II, a Valiant, a Cheyenne, and I just picked up a Ranger.  Not big enough statistical sample to draw any conclusions from, I'm afraid!

73 John


But the whole JOHNNY NOVICE'S GUIDE series is based mostly on large dollops of subjective opinion mixed with facts, so I'm sure there is wiggle room here. And you could solicit the opinions of others with direct experience with the xmtrs in question.

For example:

TOP TEN (DESKTOP) TRANSMITTERS

10 - Collins 32v3 - expensive, needs mods to de-yellify, but hey, it's a Collins

9- B&W 5100 - simple, built like a tank, heavyweight iron

8- Multi Elmac AF 68 - electrically similar to Ranger, but ugly as sin

7- Hallicrafters HT37 - fun, but prone to transformer failure

6 - Heath DX 40/DX60 - screen modulated, can be de-yellified

5- Valiant - take out the clipper, it's a classic

4 - Viking II - tank-like contruction, needs de-yellification

3 - Heath Apache - easy to de-scratch-ify

2 - DX-100 - a favorite despite some shortcomings

1 - Ranger - nice stock audio but low power is a drag, still, it's THE classic.
Logged
Ed - N3LHB
Contributing
Member
*
Offline Offline

Posts: 336


« Reply #5 on: June 15, 2005, 09:58:05 PM »

That's strange, I just restored a R-390A and Z216-1 kept going out of adjustment. I knew something was wrong because I had the slug at its maximum "down" adjustment to get a peak. Then in a few hours it would change, and I had to readjust the slug back up. I swapped it out with Z216-2, readjusted everything and the trouble stayed with the original Z216-1, which was plugged into Z216-2's slot. There are 3 Z216's in cascade, so I could swap them around a bit.
 
  Changed out the S/M first (although I suspected shorted windings on the coil) and the trouble went away. Then, a few days later the trouble came back, this time it was Z216-3 that needed adjustment. So, I replaced the S/M in Z216-3 and the R-390a has been working OKFBOM for about 3 weeks. Never would have suspected that a silver mica across a coil would ever go bad, with no real voltage across it.
 
   Hopefully no others decide to crap out, I always thought those silver micas would be the last parts to fail and last forever...
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4590



« Reply #6 on: June 16, 2005, 06:46:18 AM »

Joe - I like your list!  Maybe you oughta do the guide to xmitters.

Ed - must be something in the air lately... I've gone a quite a while without running across a bad S/M - now, the last 3 radios I've worked on have had bad ones (all Hallicrafters, coincidentally).

One sure indication that you have a bad silver mica in a resonant circuit is if you have to have the slug all the way in (or all the way out) and it still doesn't hit the peak.

73 John
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #7 on: January 30, 2006, 09:00:26 PM »

John,
I've never heard anyone call postage stamp mica caps silver mica. I wonder if they are some other metal like tin. I have seen the effects of tin many times but never saw it with silver. I thought the later silver dipped micas solved the problems of postage stamp mica caps. I have never seen a silver dipped mica cap fail.
The only problem I have seen with silver in in cavities when the silver plating flaked off in chunks killing the cavity Q. I can't ever remember it growing wiskers.
Now I need to chop one open and check....
Logged
Jim, W5JO
Member

Offline Offline

Posts: 2449


« Reply #8 on: January 31, 2006, 02:30:20 PM »

Some of those postage stamp caps are paper, some are mica.  If they have the JAN or AWS code the first dot on the top left with the arrow(s) pointo to the right tells you.    Black is Mica, Silver is paper.

The RETMA 6 dot code is different.  Most of what we deal with is the JAN code, but do be aware.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #9 on: January 31, 2006, 03:25:35 PM »

But what metal is used for the plates in the postage stamp caps?
Logged
Jim, W5JO
Member

Offline Offline

Posts: 2449


« Reply #10 on: January 31, 2006, 05:34:06 PM »

All my reference book speak about foil or aluminum foil.  The tantalum uses tantalum foil, whatever that is.

Some of the book simply state metal foil, so I am not sure, but having seen older caps apart, it looks like some form of aluminum.
Logged
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #11 on: January 31, 2006, 06:23:20 PM »

I would think it would be something that would attach to a copper lead so that is why I thought it may be tin. Tin grows wiskers like crazy. I guess I need to chop an old one open and check under the scope. 
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.14 seconds with 18 queries.